The present application relates generally to an improved data processing apparatus and method and more specifically to an apparatus and method for oversubscribing branch circuits in a data center while lowering power distribution costs.
The power consumption of information technology (IT) equipment depends on the configuration of the equipment and the workload the equipment processes. To stay within power consumption limits, IT manufacturers rate equipment for:                worst conditions, such as high temperature, high altitude, or the like;        maximum configuration, such as all processor sockets fully installed and operating, all drive bays fully installed and used, all expansion cards fully installed and operating, all DIMM slots fully installed and used, or the like; and        extreme workloads, such as:                    DAXPY, which is a combined scalar multiplication plus vector addition operation where the scalar α and the vectors x and y are a double precision α,            Linpack, which is a collection of Fortran subroutines that analyze and solve linear equations and linear least-squares problems, or            “power virus”, which finds input vectors that cause maximum dynamic power dissipation (maximum toggles) in circuits.                        
Data center and IT operators have argued for a long time that equipment seldom uses as much current/power as rated causing unnecessary power distribution cost that impedes IT growth.